public inbox for stable@vger.kernel.org
 help / color / mirror / Atom feed
From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: stable@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	patches@lists.linux.dev, Wu Bo <bo.wu@vivo.com>,
	Mikulas Patocka <mpatocka@redhat.com>,
	Mike Snitzer <snitzer@kernel.org>
Subject: [PATCH 5.4 60/94] dm verity: dont perform FEC for failed readahead IO
Date: Tue,  5 Dec 2023 12:17:28 +0900	[thread overview]
Message-ID: <20231205031526.186725927@linuxfoundation.org> (raw)
In-Reply-To: <20231205031522.815119918@linuxfoundation.org>

5.4-stable review patch.  If anyone has any objections, please let me know.

------------------

From: Wu Bo <bo.wu@vivo.com>

commit 0193e3966ceeeef69e235975918b287ab093082b upstream.

We found an issue under Android OTA scenario that many BIOs have to do
FEC where the data under dm-verity is 100% complete and no corruption.

Android OTA has many dm-block layers, from upper to lower:
dm-verity
dm-snapshot
dm-origin & dm-cow
dm-linear
ufs

DM tables have to change 2 times during Android OTA merging process.
When doing table change, the dm-snapshot will be suspended for a while.
During this interval, many readahead IOs are submitted to dm_verity
from filesystem. Then the kverity works are busy doing FEC process
which cost too much time to finish dm-verity IO. This causes needless
delay which feels like system is hung.

After adding debugging it was found that each readahead IO needed
around 10s to finish when this situation occurred. This is due to IO
amplification:

dm-snapshot suspend
erofs_readahead     // 300+ io is submitted
	dm_submit_bio (dm_verity)
		dm_submit_bio (dm_snapshot)
		bio return EIO
		bio got nothing, it's empty
	verity_end_io
	verity_verify_io
	forloop range(0, io->n_blocks)    // each io->nblocks ~= 20
		verity_fec_decode
		fec_decode_rsb
		fec_read_bufs
		forloop range(0, v->fec->rsn) // v->fec->rsn = 253
			new_read
			submit_bio (dm_snapshot)
		end loop
	end loop
dm-snapshot resume

Readahead BIOs get nothing while dm-snapshot is suspended, so all of
them will cause verity's FEC.
Each readahead BIO needs to verify ~20 (io->nblocks) blocks.
Each block needs to do FEC, and every block needs to do 253
(v->fec->rsn) reads.
So during the suspend interval(~200ms), 300 readahead BIOs trigger
~1518000 (300*20*253) IOs to dm-snapshot.

As readahead IO is not required by userspace, and to fix this issue,
it is best to pass readahead errors to upper layer to handle it.

Cc: stable@vger.kernel.org
Fixes: a739ff3f543a ("dm verity: add support for forward error correction")
Signed-off-by: Wu Bo <bo.wu@vivo.com>
Reviewed-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
 drivers/md/dm-verity-target.c |    4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

--- a/drivers/md/dm-verity-target.c
+++ b/drivers/md/dm-verity-target.c
@@ -579,7 +579,9 @@ static void verity_end_io(struct bio *bi
 	struct dm_verity_io *io = bio->bi_private;
 
 	if (bio->bi_status &&
-	    (!verity_fec_is_enabled(io->v) || verity_is_system_shutting_down())) {
+	    (!verity_fec_is_enabled(io->v) ||
+	     verity_is_system_shutting_down() ||
+	     (bio->bi_opf & REQ_RAHEAD))) {
 		verity_finish_io(io, bio->bi_status);
 		return;
 	}



  parent reply	other threads:[~2023-12-05  3:45 UTC|newest]

Thread overview: 101+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-12-05  3:16 [PATCH 5.4 00/94] 5.4.263-rc1 review Greg Kroah-Hartman
2023-12-05  3:16 ` [PATCH 5.4 01/94] driver core: Release all resources during unbind before updating device links Greg Kroah-Hartman
2023-12-05  3:16 ` [PATCH 5.4 02/94] RDMA/irdma: Prevent zero-length STAG registration Greg Kroah-Hartman
2023-12-05  3:16 ` [PATCH 5.4 03/94] PCI: keystone: Drop __init from ks_pcie_add_pcie_{ep,port}() Greg Kroah-Hartman
2023-12-05  3:16 ` [PATCH 5.4 04/94] afs: Make error on cell lookup failure consistent with OpenAFS Greg Kroah-Hartman
2023-12-05  3:16 ` [PATCH 5.4 05/94] drm/panel: simple: Fix Innolux G101ICE-L01 bus flags Greg Kroah-Hartman
2023-12-05  3:16 ` [PATCH 5.4 06/94] drm/panel: simple: Fix Innolux G101ICE-L01 timings Greg Kroah-Hartman
2023-12-05  3:16 ` [PATCH 5.4 07/94] ata: pata_isapnp: Add missing error check for devm_ioport_map() Greg Kroah-Hartman
2023-12-05  3:16 ` [PATCH 5.4 08/94] drm/rockchip: vop: Fix color for RGB888/BGR888 format on VOP full Greg Kroah-Hartman
2023-12-05  3:16 ` [PATCH 5.4 09/94] HID: core: store the unique system identifier in hid_device Greg Kroah-Hartman
2023-12-05  3:16 ` [PATCH 5.4 10/94] HID: fix HID device resource race between HID core and debugging support Greg Kroah-Hartman
2023-12-05  3:16 ` [PATCH 5.4 11/94] ipv4: Correct/silence an endian warning in __ip_do_redirect Greg Kroah-Hartman
2023-12-05  3:16 ` [PATCH 5.4 12/94] net: usb: ax88179_178a: fix failed operations during ax88179_reset Greg Kroah-Hartman
2023-12-05  3:16 ` [PATCH 5.4 13/94] arm/xen: fix xen_vcpu_info allocation alignment Greg Kroah-Hartman
2023-12-05  3:16 ` [PATCH 5.4 14/94] amd-xgbe: handle corner-case during sfp hotplug Greg Kroah-Hartman
2023-12-05  3:16 ` [PATCH 5.4 15/94] amd-xgbe: handle the corner-case during tx completion Greg Kroah-Hartman
2023-12-05  3:16 ` [PATCH 5.4 16/94] amd-xgbe: propagate the correct speed and duplex status Greg Kroah-Hartman
2023-12-05  3:16 ` [PATCH 5.4 17/94] net: axienet: Fix check for partial TX checksum Greg Kroah-Hartman
2023-12-05  3:16 ` [PATCH 5.4 18/94] afs: Return ENOENT if no cell DNS record can be found Greg Kroah-Hartman
2023-12-05  3:16 ` [PATCH 5.4 19/94] afs: Fix file locking on R/O volumes to operate in local mode Greg Kroah-Hartman
2023-12-05  3:16 ` [PATCH 5.4 20/94] nvmet: remove unnecessary ctrl parameter Greg Kroah-Hartman
2023-12-05  3:16 ` [PATCH 5.4 21/94] nvmet: nul-terminate the NQNs passed in the connect command Greg Kroah-Hartman
2023-12-05  3:16 ` [PATCH 5.4 22/94] MIPS: KVM: Fix a build warning about variable set but not used Greg Kroah-Hartman
2023-12-05  3:16 ` [PATCH 5.4 23/94] ext4: add a new helper to check if es must be kept Greg Kroah-Hartman
2023-12-05  3:16 ` [PATCH 5.4 24/94] ext4: factor out __es_alloc_extent() and __es_free_extent() Greg Kroah-Hartman
2023-12-05  3:16 ` [PATCH 5.4 25/94] ext4: use pre-allocated es in __es_insert_extent() Greg Kroah-Hartman
2023-12-05  3:16 ` [PATCH 5.4 26/94] ext4: use pre-allocated es in __es_remove_extent() Greg Kroah-Hartman
2023-12-05  3:16 ` [PATCH 5.4 27/94] ext4: using nofail preallocation in ext4_es_remove_extent() Greg Kroah-Hartman
2023-12-05  3:16 ` [PATCH 5.4 28/94] ext4: using nofail preallocation in ext4_es_insert_delayed_block() Greg Kroah-Hartman
2023-12-05  3:16 ` [PATCH 5.4 29/94] ext4: using nofail preallocation in ext4_es_insert_extent() Greg Kroah-Hartman
2023-12-05  3:16 ` [PATCH 5.4 30/94] ext4: fix slab-use-after-free " Greg Kroah-Hartman
2023-12-05  3:16 ` [PATCH 5.4 31/94] ext4: make sure allocate pending entry not fail Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 32/94] arm64: cpufeature: Extract capped perfmon fields Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 33/94] KVM: arm64: limit PMU version to PMUv3 for ARMv8.1 Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 34/94] ACPI: resource: Skip IRQ override on ASUS ExpertBook B1402CVA Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 35/94] bcache: replace a mistaken IS_ERR() by IS_ERR_OR_NULL() in btree_gc_coalesce() Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 36/94] s390/dasd: protect device queue against concurrent access Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 37/94] USB: serial: option: add Luat Air72*U series products Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 38/94] hv_netvsc: Fix race of register_netdevice_notifier and VF register Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 39/94] hv_netvsc: Mark VF as slave before exposing it to user-mode Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 40/94] dm-delay: fix a race between delay_presuspend and delay_bio Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 41/94] bcache: check return value from btree_node_alloc_replacement() Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 42/94] bcache: prevent potential division by zero error Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 43/94] USB: serial: option: add Fibocom L7xx modules Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 44/94] USB: serial: option: fix FM101R-GL defines Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 45/94] USB: serial: option: dont claim interface 4 for ZTE MF290 Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 46/94] USB: dwc2: write HCINT with INTMASK applied Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 47/94] usb: dwc3: set the dma max_seg_size Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 48/94] USB: dwc3: qcom: fix resource leaks on probe deferral Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 49/94] USB: dwc3: qcom: fix wakeup after " Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 50/94] io_uring: fix off-by one bvec index Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 51/94] perf inject: Fix GEN_ELF_TEXT_OFFSET for jit Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 52/94] pinctrl: avoid reload of p state in list iteration Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 53/94] firewire: core: fix possible memory leak in create_units() Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 54/94] mmc: sdhci-sprd: Fix vqmmc not shutting down after the card was pulled Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 55/94] mmc: block: Do not lose cache flush during CQE error recovery Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 56/94] ALSA: hda: Disable power-save on KONTRON SinglePC Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 57/94] ALSA: hda/realtek: Headset Mic VREF to 100% Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 58/94] ALSA: hda/realtek: Add supported ALC257 for ChromeOS Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 59/94] dm-verity: align struct dm_verity_fec_io properly Greg Kroah-Hartman
2023-12-05  3:17 ` Greg Kroah-Hartman [this message]
2023-12-05  3:17 ` [PATCH 5.4 61/94] bcache: revert replacing IS_ERR_OR_NULL with IS_ERR Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 62/94] powerpc: Dont clobber f0/vs0 during fp|altivec register save Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 63/94] btrfs: add dmesg output for first mount and last unmount of a filesystem Greg Kroah-Hartman
2023-12-09 17:28   ` Nathan Chancellor
2023-12-11 14:56     ` Greg Kroah-Hartman
2023-12-13 13:08       ` David Sterba
2023-12-05  3:17 ` [PATCH 5.4 64/94] btrfs: fix off-by-one when checking chunk map includes logical address Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 65/94] btrfs: send: ensure send_fd is writable Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 66/94] btrfs: make error messages more clear when getting a chunk map Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 67/94] Input: xpad - add HyperX Clutch Gladiate Support Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 68/94] ipv4: igmp: fix refcnt uaf issue when receiving igmp query packet Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 69/94] net: stmmac: xgmac: Disable FPE MMC interrupts Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 70/94] ravb: Fix races between ravb_tx_timeout_work() and net related ops Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 71/94] net: ravb: Use pm_runtime_resume_and_get() Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 72/94] net: ravb: Start TX queues after HW initialization succeeded Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 73/94] perf intel-pt: Adjust sample flags for VM-Exit Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 74/94] perf intel-pt: Fix async branch flags Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 75/94] smb3: fix touch -h of symlink Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 76/94] s390/mm: fix phys vs virt confusion in mark_kernel_pXd() functions family Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 77/94] s390/cmma: fix detection of DAT pages Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 78/94] mtd: cfi_cmdset_0001: Support the absence of protection registers Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 79/94] mtd: cfi_cmdset_0001: Byte swap OTP info Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 80/94] fbdev: stifb: Make the STI next font pointer a 32-bit signed offset Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 81/94] ima: annotate iint mutex to avoid lockdep false positive warnings Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 82/94] ovl: skip overlayfs superblocks at global sync Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 83/94] ima: detect changes to the backing overlay file Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 84/94] scsi: qla2xxx: Simplify the code for aborting SCSI commands Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 85/94] scsi: core: Introduce the scsi_cmd_to_rq() function Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 86/94] scsi: qla2xxx: Use scsi_cmd_to_rq() instead of scsi_cmnd.request Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 87/94] scsi: qla2xxx: Fix system crash due to bad pointer access Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 88/94] cpufreq: imx6q: dont warn for disabling a non-existing frequency Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 89/94] cpufreq: imx6q: Dont disable 792 Mhz OPP unnecessarily Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 90/94] mmc: cqhci: Increase recovery halt timeout Greg Kroah-Hartman
2023-12-05  3:17 ` [PATCH 5.4 91/94] mmc: cqhci: Warn of halt or task clear failure Greg Kroah-Hartman
2023-12-05  3:18 ` [PATCH 5.4 92/94] mmc: cqhci: Fix task clearing in CQE error recovery Greg Kroah-Hartman
2023-12-05  3:18 ` [PATCH 5.4 93/94] mmc: core: convert comma to semicolon Greg Kroah-Hartman
2023-12-05  3:18 ` [PATCH 5.4 94/94] mmc: block: Retry commands in CQE error recovery Greg Kroah-Hartman
2023-12-05  3:53 ` [PATCH 5.4 00/94] 5.4.263-rc1 review Guenter Roeck
2023-12-05  4:33   ` Greg Kroah-Hartman
2023-12-05 11:09 ` Jon Hunter

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20231205031526.186725927@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=bo.wu@vivo.com \
    --cc=mpatocka@redhat.com \
    --cc=patches@lists.linux.dev \
    --cc=snitzer@kernel.org \
    --cc=stable@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox